Abstract: A method for predicting side effects of a combination of drugs administered concurrently includes training a multi-modal cell complex neural network (MCXN) on a dataset. The MCXN includes nodes representing the drugs and proteins, pair-wise relationships between nodes representing interactions between pairs of drugs and/or proteins, and k-wise relationships between the nodes representing interactions between k drugs and/or proteins, where k¿2. The training dataset includes a list of drugs, a list of proteins, and pharmacological information about the drugs in the list of drugs and proteins in the list of proteins. A specification of the combination of at least three drugs to be administered concurrently is input to the MCXN which predicts probabilities that administering the combination of drugs concurrently results in potential side effects. It also predicts both frequencies of the potential side effects and severities of the potential side effects.

Abstract: A method for object recognition from point cloud data acquires irregular point cloud data using a 3D data acquisition device, constructs a nearest neighbor graph from the point cloud data, constructs a cell complex from the nearest neighbor graph, and processes the cell complex by a cell complex neural network (CXN) to produce a point cloud segmentation or a point cloud classification using geometric message passing schemes to implement deep learning protocol in the CXN. The point cloud segmentation may include an object classification label for each point in the point cloud, and/or a classification label identifying an object in the point cloud.